* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Multiple sites of spike initiation in a single dendritic
Action potential wikipedia , lookup
Neural oscillation wikipedia , lookup
Long-term depression wikipedia , lookup
Neuroregeneration wikipedia , lookup
Microneurography wikipedia , lookup
Neuromuscular junction wikipedia , lookup
Neuropsychopharmacology wikipedia , lookup
Premovement neuronal activity wikipedia , lookup
Development of the nervous system wikipedia , lookup
Neurotransmitter wikipedia , lookup
Central pattern generator wikipedia , lookup
Biological neuron model wikipedia , lookup
Node of Ranvier wikipedia , lookup
End-plate potential wikipedia , lookup
Anatomy of the cerebellum wikipedia , lookup
Molecular neuroscience wikipedia , lookup
Pre-Bötzinger complex wikipedia , lookup
Stimulus (physiology) wikipedia , lookup
Axon guidance wikipedia , lookup
Dendritic spine wikipedia , lookup
Neuroanatomy wikipedia , lookup
Multielectrode array wikipedia , lookup
Optogenetics wikipedia , lookup
Electrophysiology wikipedia , lookup
Activity-dependent plasticity wikipedia , lookup
Neural coding wikipedia , lookup
Feature detection (nervous system) wikipedia , lookup
Holonomic brain theory wikipedia , lookup
Apical dendrite wikipedia , lookup
Nervous system network models wikipedia , lookup
Channelrhodopsin wikipedia , lookup
Synaptic gating wikipedia , lookup
Nonsynaptic plasticity wikipedia , lookup
Chemical synapse wikipedia , lookup
Caridoid escape reaction wikipedia , lookup
Synaptogenesis wikipedia , lookup
316 Bra& Research, 82 (1974) 316-321 © Elsevier Scientific Publishing Company, Amsterdam - Printed in The Netherlands Multiple sites of spike initiation in a single dendritic system RONALD L. CALABRESE* AND DONALD KENNEDY Department of Biological Sciences, StanJbrd University, Stanford, Calif. 94305 (U.S.A.) (Accepted September 17th, 1974) Crayfish multisegmental tactile interneurons (MTIs) provide a unique and convincing illustration that multiple axonal spike-initiating zones exist in individual central neuronsL The implications of these findings for other neuronal systems have largely been ignored; the association of each initiating zone with a separate ganglionic neuropile allows the phenomenon to be dismissed as a curiosity found only in invertebrates with ladder-like nerve cords 1. In all cases of central neurons in which multiple spike-initiating zones have been demonstrated within a single dendritic arbor (e.g., Purkinje ceils of alligator cerebellumS, 9, pyramidal cells of cat hippocampus, crab eye stalk motor neuronslZ, 1~, and crayfish fast flexor m o t o r neurons15), the dendritic spikes contribute only subthreshold excitation to a main axonal spike initiating zone. Such dendritic spikes, which do not propagate into the axon because they must traverse regions of low safety factor, serve the same function as classical postsynaptic potentials in neurons with large dendritic trees (e.g., Purkinje cells s,9) or small (e.g., crab eye stalk motor neuronsl2, ~3) diameter dendrites. There are indications in some neurons, however, that dendrite spikes may have access to the axon. In immature cat neocortical neurons and the hippocampal neurons of adult cats in which seizures have been induced, dendritic spikes often actively invade the somalL Some dendritic spikes in Purkinje cells are conducted to the axon without failure s,9 and in fish oculomotor neurons, firing level differences indicate that some spikes that actively invade the axon may be initiated at sites distant from the main axonal site v. We here report that the crayfish M T I s with bilateral receptive fields have multiple spike-initiating zones within a single dendritic arbor, and that each has direct access to the axon. Moreover, we show that these spike-initiating zones are so widely separated that they cannot be jointly influenced by subthreshold EPSPs. The abdominal nervous system of the crayfish (Procambarus clarkii) was isolated from the animal except for its connection to the tactile hair receptors of the telson via the fourth and (sometimes) the fifth roots of the sixth abdominal ganglion. These roots are almost exclusively sensory and represent potent excitatory pathways for a number * Present address: Department of Molecular Biology, University of California, Berkeley, Calif. 94720, U.S.A. SHORT COMMI [NICATIONS 3 17 j<..5 t B CJ lilIDLIN[ 2 MIDLINE Y Ax al S Ii "X,s Fig. 1. A: the bottom traces are intracellular recordings from the contralateral dendrite of interneuron C. The top traces are simultaneous extracellular recordings from the axon of interneuron C (spikes marked with arrows) in the interganglionic connectives. The extracellular electrode also records the spikes associated with other interneurons beside C. In 1 the contralateral fourth root has been stimulated, while in 2 the ipsilateral homolog has been stimulated. Ati (the difference in occurrence time of the interneuron spike at the two recording sites, measured from the foot of rising phase of the spikes, for ipsilateral stimulation) equals 0.5 msec, while 2xtc (the corresponding occurrence time difference for contralateral stimulation) equals 0.9 msec. The latency and the intradendritically recorded amplitude of interneuron C spikes are both variable. Calibrations: 5 mV and 2.0 msec. B: drawing of a cobalt sulfide impregnation of interneuron C within the sixth abdominal ganglion (ventral view). The cell was filled by dipping its isolated axon in 300 m M CoCI~ overnight and then developing with (NH4)2S. al, ae, and b are the main dendritic branches of interneuron C within the sixth ganglion. S and Ax are soma and axon, respectively. R4r and R41 are right and left fourth roots, respectively. Calibration: 1 ram. C: stick models of interneuron C within the sixth ganglion. Y is the site of extracellular recording, and X is the site of intracellular recording for the experiment described in A. I, h, and Ic are hypothetical sites of spike-initiating zones. In 1 there is a single spike-initiating zone, while in 2 there are separate spike-initiating zones for the two sides of the cell. All other symbols are the same as in B. 318 SHORT COMMUNICATIONS of MTIs. Suction electrodes were used for electrical stimulation of these roots and for extracellular recording from the ventrolateral surface of the desheathed interganglionic connectives that contain the axons of the MTIs. The sixth abdominal ganglion was desheathed and probed from the ventral surface with 3 M KCl-filled micropipettes having resistances in the range of 40-60 M R when measured in the physiological saline. With such microelectrodes, it is possible to penetrate dendrites of the MTIs and to record both subthreshold and spike activityL We were encouraged to look for multiple spike-initiating zones because of the observation of Kennedy and Mellon 4 that, in MTIs with bilateral receptive fields, the apparent voltage thzeshold for firing differed for synaptic input from the two sides. Synaptic input from roots ipsilateral to the dendritic penetration generated spikes arising from large EPSPs, whereas contralateral root stimulation generated spikes that arose from very small EPSPs 4. We have repeated this observation for an identified MTI, interneuron C (Fig. 1A) 14. This finding could be explained by assuming that the two sides have separate spike-initiating zones, and that the microelectrode is closer to the one associated with the larger compound EPSP. We tested this hypothesis by employing a two-point recording technique similar to the one used by Kennedy and Mellon to demonstrate that MTIs have separate spike-initiating zones for each ganglion in which they receive input 4. In this case we recorded extracellularly from the axon of an MTI in the interganglionic connectives and intracellularly from its dendrites in the neuropile of the sixth ganglion. Fig. 1B is a drawing of a cobalt sulfide impregnation ofinterneuron C. There are three main dendritic branches (two ipsilateral to the axon and one contralateral labeled al, a2 and b, respectively) and we reasoned that each might have its own spike-initiating zone. However, since we were primarily concerned with the comparison of synaptic input from bilateral pairs of roots, we did not consider the added complexity of more than one spike-initiating zone on a particular side, although this possibility does not alter the logic of our analysis. Anatomical and physiological evidence will be presented elsewhere (Calabrese, in preparation) that in the abdominal nerve cord tactile afferent fibers never cross the midline. Therefore we can be sure that stimulation of a sensory root on one side results in monosynaptic input confined to MTI dendritic branches of that same side. Fig. 1C gives two alternative schematic models of interneuron C with recording electrodes in place (Y is the extracellular recording site, and X is the intracellular recording site). Model 1 (Fig. 1C1) assumes the existence of a single spike-initiating zone at the arbitrary point I. Under these circumstances, no matter where the microelectrode is located in the dendritic arbor, the difference in the time of occurrence at X and Y of a spike evoked by ipsilateral stimulation (Ati) will be equal to the corresponding difference in the time of occurrence at X and Y of a spike evoked by contralateral stimulation (Ate). Model 2 (Fig. 1C~) assumes that there are at least two spikeinitiating zones (one for each side), for example at the arbitrary points I~ and Ie. In this case, if the microelectrode tip is located between the two initiating zones, for example at X, then Ah will not equal Ate. (The amount by which Ati deviates from Ate will increase as X is made closer to Ie, regardless of whether X is to the left or to the 319 SHORT COMMUNICATIONS A1 i2 I - _ ~ I . _eIj J I I I B~ 2 Fig. 2. Interaction of subthreshold EPSPs between two spike-initiating zones of interneuron C. Recordings are from same penetration as Fig. 1A. A: the bottom traces are intracellular recordings from the contralateral dendrite of interneuron C and the top traces are simultaneous extracellular recordings from the axon of interneuron C (spikes marked with arrows) in the interganglionic connectives. At equals 0.5 msec for both A2 and A3. B: traces are as described in A. At equals 0.9 msec for both B2 and Bs. See text for explanation of figure. Calibrations: 5 mV and 2.0 msec. right o f I.) Even if there are separate spike-initiating zones associated with the two synaptic inputs tested, At~ can equal Ate if the microelectrode penetration is at a point on a c o m m o n o r t h o d r o m i c conduction path for spikes initiated at the two zones, or under certain other fortuitous circumstances (e.g., if the microelectrode is in branch al and spikes are initiated at I~ and Ie by the inputs tested). If, however, it is possible to show that Atg does not equal Ate for the two inputs tested, then it is a logical necessity that separate spike-initiating zones exist for the two inputs. In order to maximize our chances o f detecting a difference between Ati and Ate, we penetrated interneuron C in a dendritic branch located contralateral to its axon. Interneuron C was identified by its receptive field, the position o f its axon in the interganglionic connectives and its characteristic firing pattern. We confirmed that we had penetrated a contralateral dendritic branch by showing (1) that an antidromic spike could be evoked by electrical stimulation o f the interganglionic connective opposite the intracellular recording site, and (2) that each intracellular spike was followed by impulses in a single unit recorded intracellularly in the interganglionic connective opposite the microelectrode. We then c o m p a r e d the At's associated with shocks to the two fourth roots. The results o f such an experiment are shown in Fig. 1A and lB. N o t e that Ati (0.5 msec) does not equal Ate (0.9 msec), indicating that the cell has separate spike-initiating zones associated with the two populations o f fourth r o o t 320 SHORT COMMUNICATIONS afferents. A clude estimate of the average conduction velocity within the dendritic arbor can be obtained by dividing the distance from X to Y (in this case about 2.5 mm) by Ate. This yields a value of 2.6 m/sec, in the same velocity range as axonal conduction in the same neurons. In two other similar experiments (one on interneuron C and the other on an unidentified MTI) the difference between Ati and Ate was 0.3 msec for each, which is similar to the 0.4 msec difference for the experiment of Fig t. Fig. 2 shows the result of another experiment on the same cell, designed to detect subthreshold interactions between synaptic input to the two sides. Fig. 2At shows the EPSP associated with a just-subthreshold shock to the fourth root contralateral to the axon. In Fig. 2A2 the cell is brought to threshold by adding the smallest effective simultaneous shock to the ipsilateral fourth root. Note that the At associated with the spike is 0.5 msec, indicating that it is initiated at the zone ipsilateral to the axon. In Fig. 2A8, the subthreshold stimulus to the contralateral fourth root is removed, with no change in the firing of the cell. Fig. 2B shows that the converse experiment produces equivalent results. We conclude that the two spike-initiating zones are incapable of significant subthreshold interaction. Since ipsilateral subthreshold root shocks do result in recordable EPSPs contralaterally, the microelectrode penetration must be between the two spike-initiating zones but much closer to the contralateral one. Using similar techniques, we tested (in another MTI) the possibility that input from fourth and fifth root from the same side might influence different spike-initiating zones; the results were negative. This is not surprising, since fourth and fifth root afferents project to the same neuropilar areas and, therefore, probably terminate on the same dendritic branch of a given postsynaptic cell. We have to date not performed this experiment using homolateral pairs of roots which terminate in disparate neuropilar areas. Unlike the dendritic spikes observed in other neurons, those in crayfish MTIs have direct access to the axon. Multiple spike-initiating zones exist within a single dendritic arbor, and the interaction of subthreshold EPSPs between them is limited. If these properties are found by similar tests to occur in other neurons, it will challenge the widely accepted view that all inputs upon a postsynaptic neuron are integrated at a single point by subthreshold potentials. Crayfish MTIs are known to receive powerful postsynaptic inhibition in the sixth ganglion resulting from giant fiber discharge 5& Our results indicate that such inputs would have to be widely distributed in the dendritic tree of such an MTI in order to act against all its excitatory inputs. Conversely, the existence of multiple spike-initiating zones within a single dendritic arbor, each associated with a different set of excitatory synaptic inputs, makes it possible for postsynaptic inhibition to be directed against specific sources of excitation. Supported by an N.S.F. Predoctoral Fellowship and a P.H.S. Predoctoral Traineeship (GM 00712-16) to R.L.C. and by N.I.H. Grant NB 02944 to D.K. We thank Pamela Decker for assistance with the cobalt preparations, Alan Gelperin and DeForest Mellon for critical reviews. SHORT COMMUNICATIONS 321 1 HORRIDGE,G. A., Interneurons, Freeman, San Francisco, Calif., 1968, pp. 96-126. 2 HUGHES, G. M., AND WlERSMA, C. A. G., Neuronal pathways and synaptic connections in the abdominal cord of the crayfish, J. exp. Biol., 37 (1960) 291-307. 3 KENNEDY,D., AND PRESTON,J. B., Activity patterns of interneurons in the caudal ganglion of the crayfish, J. gen. Physiol., (1960) 655-670. 4 KENNEDY,D., AND MELLON,DEF., Synaptic activation and receptive fields in crayfish interneurons, Comp. Biochem. Physiol., 13 (1964) 275-300. 5 KENNEDY,D., CALABRESE,R., AND WINE, J. J., Presynaptic inhibition: primary afferent depolarization in crayfish neurons, Science, in press. 6 KRASNE,F. B., ANDBRYAN,J. S., Habituation: regulation through presynaptic inhibition, Science, 182 (1973) 590-592. 7 KR~EBEL,M. E., BENNETT,M. V. L., WAXMAN,S. G., AND PAPPAS,G. D., Oculomotor neurons in fish: electrotonic coupling and multiple sites of spike initiation, Science, 160 (1969) 520-524. 8 LLIN~.S,R., NICHOLSON,C., FREEMAN,J. A., AND HILLMAN,D. E., Dendritic spikes and their inhibition in alligator Purkinje cells, Science, 160 (1968) 1132-1135. 9 LUNGS, R., AND NICHOLSON, C., Electrophysiological properties of dendrites and somata in alligator Purkinje cells, J. Neurophysiol., 34 (1971) 532-551. 10 PITMAN, R. M., TWEEDLE, C D., AND COHEN, M. J., Branching of central neurons: intracellular cobalt injection for light and electron microscopy, Science, 176 (1972) 412-414. 11 PURPURA,D. P., Comparative physiology of dendrites. In G. C. QUARTON,T. MELNECHUCKAND F. O. SCHMITT(Eds.), The Neurosciences: A Study Program, Rockefeller University Press, New York, 1967, pp. 372-393. 12 SANDEMAN,D. C., The site of synaptic activity and impulse initiation in an identified motoneurone in the crab brain, J. exp. Biol., 50 (1969) 771-784. 13 SANDEMAN,D. C., Integrative properties of a reflex motoneuron in the brain of the crab Carcinus maenas, Z. vergl. Physiol., 64 (1969) 450-464. 14 ZUCKER,R. S., KENNEDY,D., ANDSELVERSTON,A., Neuronal circuit mediating escape responses in crayfish, Science, 173 (1971) 645-650. 15 ZUCI(ER, R. S., Crayfish escape behavior and central synapses. III. Electrical junctions and dendrite spikes in fast flexor motoneurons, J. Neurophysiol., 35 (1972) 638-651.